M. Herman[2]

International Atomic Energy Agency, Vienna, Austria

Lectures given at the
Workshop on Nuclear Reaction Data and
Nuclear Reactors: Physics, Design and Safety
Trieste, 25 February — 28 March 2002

LNS0520002

The status and contents of the Reference Input Parameter Library (RIPL) are summarised. This input library provides an extensive database of model parameters for theoretical calculations of nuclear reactions. It was developed to facilitate use of reaction codes and in­crease the accuracy of theoretical predictions.

1 Introduction

Increased use of nuclear reaction theory for predicting cross sections, spectra, and angular distributions, as required for a large variety of applications, is an important trend in the evaluation of neutron and charged-particle nuclear data. The model codes offer important advantages such as ensuring internal consistency of the data by preserving the energy balance and the coherence of the partial cross sections with the total or the reaction cross sections. These features are essential for transport calculations. In addition, theoretical cal­culations represent the only approach that can fill gaps in the experimental results and predict data for unstable nuclei. Nuclear astrophysics and the design of Accelerator Driven Systems are typical applications that depend strongly on theoretical calculations.

With recent formulation of nuclear reaction models (triple-integral form of the statistical model [1], quantum mechanical Multistep Direct and Mul­tistep Compound [2, 3, 4]) and existing approaches to direct reactions, nu­clear reaction theory is believed to be in a position to meet most of the requirements for practical applications. The major sources of uncertainty are, the input parameters needed to perform theoretical calculations, in­cluding nuclear masses, deformations, nuclear levels and their decay charac­teristics, y-strength functions, neutron resonances and level densities, optical model parameters, and fission barriers. The IAEA has addressed these needs through a Coordinated Research Project on the Reference Input Parameter Library (RIPL), which involves the difficult task of collecting, evaluating and recommending the vast amounts of various nuclear parameters. RIPL is targeted at users of nuclear reaction codes and, in particular, at nuclear data evaluators. The first phase of the project was completed in 1999, with the production of a Starter File and related documentation [5]. A second phase of the project was initiated in 1999 to test the RIPL-1 database and produce interfaces between RIPL and commonly used nuclear reaction codes.

Substantial improvements and extensions to the original database have been made, resulting in a more accurate and reliable library. All files se­lected for RIPL-2 have been prepared in the unified RIPL-2 format, which facilitates their use in the reaction codes. The RIPL-2 library is expected to be released in July 2002. The contents of the RIPL-2 library are out­lined below, with possible improvements that could be made to the current database through new measurements at the SNS.

2 Contents of RIPL-2

2.1 Segment 1: MASSES

The mass segment contains basic ground state properties of nuclei, along with two theoretical predictions of masses and deformations. On the basis of the Hartree-Fock-Bogolubov (HFB) theory, a 10-parameter Skyrme force, along with a 4-parameter delta-function pairing force (with blocking for odd nuclei) and a 3-parameter Wigner term, was fitted to all 1888 measured masses of nuclei with N and Z > 8. The second file contains predictions obtained within the Finite Range Droplet Model (FRDM) [6]. The atomic mass excesses and nuclear ground-state deformations are tabulated for 8979 nuclei ranging from 16O to A=339. These calculations are based on the finite-range droplet macroscopic model and the folded-Yukawa single-particle microscopic model. The most recent evaluated experimental masses by Audi and Wapstra [7] are included as a separate column. A third possibility is provided by a subroutine implementing the Duflo-Zuker formula [8] for nu­clear masses in which, the nuclear Hamiltonian is separated into a monopole term and a residual multipole term. The monopole term is responsible for saturation and single-particle properties, and is fitted phenomenologically while the multipole part is derived from realistic interactions. The latest version of the mass formula made of 10 free parameters reproduces the 1950 experimental masses above 4He with an rms error of 574 keV.

RIPL-2 also provides natural abundancies according to the Wallet Cards and HFB matter densities. The data are necessary for calculation of optical model parameters within the semi-microscopic approach (code MOM) in segment 4.

Greenwood and co-workers have developed total absorption gamma-ray spectrometry (TAGS) in order to determine the intensity distributions of a wide range of P— emitting radionuclides (Greenwood et al, 1994; Helmer et al, 1994; Greenwood et al, 1996; Greenwood et al, 1997). The TAGS system consists of a NaI(Tl) detector with a deep axial counting well to allow almost complete summing of the gamma-ray cascades from the source. A Si detector is also located in the well so that P—particle- gated coincidence gamma rays can be measured in addition to singles gamma-ray spectra. Spontaneous fission of 252Cf generated the fission-product radionuclides, and a He gas-jet transported these nuclides to a mass separator. The selected fission — product mass fraction was collected on a tape for a pre-set time, before moving to the Si detector in the well of the TAGS system. Although only relative P—intensity distributions are obtained from the TAGS gamma-ray cascade-summed spectral data, the detector system can be used in 4ny-P — coincidence mode to obtain the P— branching intensity to the ground state; the measurement of this key parameter provides a simple means of converting the relative P—intensity distributions to absolute values.

The resulting singles spectra were compared with simulated spectra derived from the evaluated decay data to be found in Nuclear Data Sheets (Bhat, 1992): examples are shown in Figs. 26 and 27 for 138gCs and 138mCs, respectively. Measured and
calculated spectra for 138gCs are in good agreement below ~3200 keV, but the simulated spectrum exhibits serious deficiencies in the postulation of beta-particle emission probabilities above this energy. Additional pseudolevels have been introduced to the decay scheme in order to achieve the good fit shown in Fig. 28 (see also Table 11); these additions and increased populations to other high-energy levels increase the beta-particle emission probabilities above 3200 keV from 2.65% to 5.9%. Similar procedures can be applied to the spectrum of 138mCs (Fig. 29), with the addition of a considerable number of pseudolevels above 2500 keV as listed in Table 12.

The TAGS method of spectral measurement and analysis would appear to be an extremely powerful means of modifying inadequate decay-scheme data in a quantitative manner. This approach is very welcome, and should be encouraged further in order to eliminate serious gaps in our detailed knowledge of the discrete decay data of many important short-lived fission products.

NDS: Nuclear Data Sheets, 69(1993)69. P: placement of pseudolevel.

“Contribution to the total 138mCs decay is obtained by multiplying Pp by 0.19. NDS: Nuclear Data Sheets, 69(1993)69.

P: placement of pseudolevel.

There are five primary areas of industrial heat applications: food processing, paper industry, chemical industry, petroleum and coal processing, and primary metal industries. Relative use of process heat in these industries is shown in Table VII for two developed countries, the U. S. (1994) and Germany (1989).

Industrial process heat is mainly used in the form of steam at appropriate temperature and pressure conditions. The demand is usually steady and there is no seasonal variation and hence quite suitable for supply by nuclear power. The only problem is that the source must be nearby as heat loss in transit is considerable.

There are three cases of commercial use of nuclear process heat in Canada, Germany and Switzerland. This is shown in Table VIII. The application to the heavy water production facility in Bruce, Canada was the largest use of nuclear process heat and it has operated very successfully for over 20 years. The six other industries the Bruce complex provided process heat were plastic film manufacturing, ethanol plant, apple juice concentration plant, alfalfa dehydration, cubing and pelletizing plant, a greenhouse, and an agricultural research facility.

The total potential market of industrial process heat is large and of the same order of magnitude as district heating. It is estimated to be between 240 GWt and 2900 GWt2. However, the demand in terms of size varies; some 50% of the users need less than 10 MWt, 40% need sizes from 10 to 50 MWt, and only 10% need sizes greater than 50 MWt3. Very few need a large amount of process heat as in the Bruce example in Canada. The market is also very competitive as small fossil fuel units can provide the needed steam.

A. L. Nichols*

International Atomic Energy Agency, Nuclear Data Section,
Department of Nuclear Sciences and Applications,
Vienna, Austria

Lectures given at the
Workshop on Nuclear Reaction Data and
Nuclear Reactors: Physics, Design and Safety
Trieste, 25 February — 28 March 2002

LNS0520003

Abstract

A sound knowledge of the time-dependent energy release resulting from the decay of the radioactive nuclides formed in the reactor core is extremely important in formulating safe procedures for the operation of nuclear facilities and the handling of irradiated fuel. Accurate estimates of this resulting decay heat are needed for a wide range of applications, including safety assessments of all types of nuclear plant, the handling of fuel discharges, the design and transport of fuel-storage flasks, and the management of the resulting radioactive waste. More specifically, the nuclear power community must ensure accurate and reliable calculations of the decay heat of irradiated fuel in order to maintain credibility and confidence in the safe and reliable performance of the various nuclear fuel cycles.

Neutron cross sections, fission yields and radionuclidic decay data represent the input to the summation calculations used to determine the release of decay heat over an extended period of time after reactor shutdown (i. e., following termination of neutron-induced fission). Nuclear data requirements for these summation calculations have been assessed, and a summary is given of their status. Associated uncertainties are examined, and specific inadequacies explored.

1. INTRODUCTION

Neutron-induced fission within the fuel of a reactor core, and the subsequent conversion of mass to energy constitutes a principal means of generating power. The resulting energy arises from the following phenomena:

kinetic energies of the fission products and neutrons;

prompt gamma radiation from highly-excited fission fragments, including short-lived isomeric states;

gamma and beta energy released through the delayed natural decay of the various radioactive products, particularly the fission products.

The last energy source contributes approximately 8% to 12% of the total energy generated through the fission process, and is commonly referred to as “decay heat”.

The author has focused on the decay heat that continues to be generated after the fission process has been terminated; the fission process is not considered directly, and the reader is referred to a number of dedicated publications on this phenomenon (Gonnenwein, 1991; Wagemans, 1991; Denschlag, 1997). The prompt sources of energy decline rapidly when a reactor is shutdown, but radioactive decay continues to heat the reactor core. Hence, coolant operation needs to be maintained after termination of the fission process, on the basis of reliable decay-heat calculations. Decay heat varies as a function of cooling time, and can be determined in theory from known nuclear data, based on computations of the inventory of the resulting radionuclides (primarily fission products, and actinides) created during the fission process and after reactor shutdown, and their radioactive decay characteristics.

Decay heat has been reviewed in detail by others from a technical perspective and also through the use of decay-heat equations as standards (ANS, 1979; Tobias, 1980; GNS, 1990; Dickens et al, 1991; Tasaka et al, 1991); the reader is referred to these publications for authorative assessments of the analytical procedures. Rather, the author has focused on the basic nuclear data of relevance to the decay heat generated in fission power-reactor systems, including the means of defining the initial radioactive inventory of controlled nuclear fission and other modes of decay within the core.

Cross-section, fission-yield and decay-data libraries are maintained for national and international usage. While this article avoids recommending specific sources of such data, Appendix A provides the reader with a brief summary of the means of accessing the most relevant data files via the acquisition of CD-ROMs or through the Internet. The latter method has become increasingly powerful, and provides the user with an extremely rapid route to virtually all of the highest quality nuclear data.

A brief summary is given of some of the on-line services (mainly in Europe and the USA), user-friendly PC programs and CD ROMs available to search for, display and retrieve nuclear data:

— evaluated nuclear data files — include fission yields, nuclear structure, mass, radioactive decay and cross-section data,

— scientific works published recently in the literature — e g., NSR (Nuclear Science References, a computer file of indexed references maintained by NNDC, Brookhaven National Laboratory, USA); INIS (International Nuclear Information System) operated by the International Atomic Energy Agency, Vienna, Austria; references system for the NUBASE evaluation (Atomic Mass Data Centre, Orsay, France), with regular updates accessible through the Internet.

Each of these tools may be rapidly used by reactor physicists to access evaluated data (e. g., ENSDF, AME, NUBASE, ENDF/B and JEF), and by nuclear data evaluators as a source of information on new measurements. Such on-line services and PC programs are also extremely useful for research and educational purposes.

The list of accessible data libraries outlined below is not complete — there are numerous special purpose files dedicated to more specific applications that have not been included (e. g., NUCLEIDE — CD-ROM reference files of decay data for radiometrology (Be et al, 1996); BANDRRI — CD-ROM reference files of decay data for dosimetry applications (Los Arcos et al, 2000); CD-ROM containing gamma-ray spectrum catalogues based on the studies of Heath (1974), and expanded by Helmer et al (2000)). Many of these specialised data files can be applied to the needs of nuclear medicine, dosimetry and detector calibration, and are accessible through the Internet.

The issue of health and safety aspects and the broader one of the environmental impact of the decommissioning process is far reaching and it may only be summarized.

It is convenient to distinguish at least three aspects:

1. Occupational safety, or safety of the workers directly involved in the decommissioning activities

2. Public safety, or the safety of the population surrounding the plant in decommissioning, excluding therefore those who may be affected by the waste disposal process

3. Environmental protection, including those aspects that are not directly related to human health

The first aspect is probably the most significant. Decommissioning is a very labour intensive activity and workers will be in contact with radioactive and other toxic wastes. However, all the means of the plant are still available to reduce the worker doses to the minimum and, while individual doses will always be below acceptable levels, an ALARA (As Low As Reasonably Achievable) analysis of single activities and process could reduce the cumulative occupational doses to values below a few years of plant operation for the entire decommissioning process (few hundreds of man-Sv).

Risks to the public are extremely low in comparison with those associated with plant operation. Radioactive inventory available for release to atmosphere or water bodies is a very small fraction of the previous ones. In general the most dangerous situations are associated with large fires in contaminated areas, breaks in tanks with large inventories of liquid radwaste, drop of contaminated loads. All these situations, however, in general would not even require the activation of an emergency plan.

The environmental issues, finally, are treated in a Environmental Impact Assessment (EIA), which is now required in Europe by a Directive of the European Union. Currently a lot of work is undergoing for a better harmonization of such assessments among the member countries. It is also needed to discuss the interaction between the EIA, presented generally to the Ministry of Environment, and the safety assessments that are presented to the Nuclear Regulatory Bodies. As in other EIA, the assessment would present an overall broad view on all interactions of the decommissioning with various environmental matrixes, and would include aspects not included in the Safety Report such as, for example, those related to site restoration, impact of material (radioactive and non radioactive) transports, disturbance to the local flora and fauna, etc.

With the methods described in the previous chapter, we can now calculate all matrix elements of the Hamiltonian H or the norm matrix N in a basis, where the radial dependence of the relative wave function of two fragments is given by rLrel exp(-вг2) for an relative orbital angular momentum Lrei. Choosing a model space and solving the generalized eigenvalue problem de­termines the eigenfunctions rv of eqs. (2.15) and (2.16). In case of a bound state calculation, these wave functions, given as an expansion in terms of Gaussians, are a variational solution. For a scattering calculation more has to be done.

Nuclear-charge distributions describe the dispersion of yields with mass and atomic numbers (A and Z) of ~1000 primary fission products from each of the many fission reactions (yields are for the fission products after prompt-neutron emission and before beta decay). However, only a small fraction of the yields have been measured, and theoretical models are not sufficiently advanced to give reliable yield estimates. Therefore, two empirical models have been proposed that include mathematical functions derived from the available experimental data (Wahl, 1988).

While the ZP and A’P models give different perspectives on the nuclear-charge distribution from fission, comparisons of the results have been helpful in improving both models. The ZP model treats the dispersion of fractional independent yields [FI] with Z for each A of primary fission products; the A’P model considers the dispersion of independent yields [IN] with A’ for each Z of primary fission products, where A’ is the average mass number of the precursor primary fragments that give products with A by prompt-neutron emission (A’ = A+ VA).

Experimental independent yields can be supplemented by yields calculated from the ZP or A’P models for the fission reaction of interest, prior to dividing amongst the isomeric states (Madland and England, 1977). The spins of the states need to be defined for both treatments, and the isomeric cumulative yields can be calculated by summation of the independent yields based on the spins and branching fractions to the isomeric states.

(a) Zp model

The ZP model has been used for charge distribution calculations in the two major fission yield evaluations (US ENDF/B-VI and UKFY2 (and UKFY3)). ZP-model parameters are defined as FZ, FN, az and AZ, and are either constant or linear functions of A’ in each of the regions considered (as shown in Fig. 11). Fractional independent yields are expressed in terms of these parameters:

FI(A, Z) = (0.5) [F(A)] [N(A)] [erf(V) — erf—

Z (A) — Zp (A) + 0.5
V2 oy(A)

Z (A) — Zp (A) — 0.5
V2 [oz (A1)]

in which Zp (AH ) = A’H [ZF / AF ] + AZ(A’H )

O. Fig. 10 Various Gaussian curves fitted to mass distributions of fission processes:

249 252

thermal-neutron fission of Cf, and spontaneous fission of Cf

98 A. L. NlChOlS

and Zp (Al ) = A[Zp / Ap ] — AZ(A’Hc)

where ZF and AF are the atomic number and mass number of the fissioning nuclide, Ah and Al are the mass numbers for the heavy and light fission peaks respectively, A’Hc = Ap — A’l, and

N(A) is the normalisation factor to achieve ЦРІ) = 1.00 for each A, which has to be implemented because P(A) destroys the inherent normalisation of the Gaussian distributions; values of N(A) seldom deviate by more than 10% from unity.

All of the parameters AZ(A/), <jz (A/), PZ (A’) and PN (A/) depend on A’, and the

region in which A’ falls. Apart from distributions close to symmetry, all of the parameters are constant except AZ, which has a small negative slope (AZ is the displacement of ZP from an unchanged charge distribution):

AZ = {Zp — A'[Zf /Af]}h = {A'[Zf /Af] — Zp}L

Near to symmetry, the width parameter [oz] changes abruptly twice to a lower value, the even-odd proton and neutron factors [FZ and FN] equal 1.0, and the AZ function undergoes a zig-zag transition from positive values for light fission products to negative values for heavy fission products. The ZP model has been modified to include parameter slopes vs. A’ and slope changes in the wing regions.

Values of the ZP-model parameters have been determined for each of twelve fission reactions (229Th thermal, 232Th fast, 233U thermal, 235U thermal, 238U fast, 238U 14

MeV neutrons, 238Np thermal, 239Pu thermal, 241Pu thermal, 242Am thermal, 249Cf thermal, and 252Cf sf) by the method of least squares. However, all of the model parameters can only be determined by this method for the thermal fission of 235U. Other fission reactions require the following equation to estimate ZP-model parameter values [PM]:

PM = P(1) + P(2)(Zf — 92) + P(3)(Af — 236) + P(4)(E* — 6.551)

Fig. 11. Zp functions for thermal fission of 235U

Solid lines calculated by L. S., red. %2 = 3.6(0.7) Dashed lines from systematics, red. %2 = 6.6(0.6)

where E* is the excitation energy above the ground state of the fissioning nuclide. P(i) values were calculated by the method of least squares from the PM values that could be determined by least squares for individual fission reactions. Fig. 12 shows the adopted parameter values as points and the derived functions as lines.

Checks have been made on the validity and usefulness of the ZP-model parameters obtained from the PM calculations for each of the 12 fission reactions investigated. The reduced X values were within a factor of approximately 2 of those determined by the method of least squares with variation of as many parameters as possible. Further details of these studies can be found in IAEA-CRP (2000).

b) A’p model

Studies are underway to determine whether a simplified A’P model can be used with only five Gaussian functions to represent the element yields, rather than the many needed for the association of one model parameter with each element yield (IAEA — CRP, 2000). Fig. 13 shows the results of calculations for the thermal-neutron fission of 235U close to symmetry (Z = 42-50).

The data include independent yields for individual elements or for complementary element pairs when the data are available from radiochemical and on-line mass — separator measurements. Solid-black symbols represent 43Tc yields, while open symbols represent the yields for heavy fission products. The Gaussian width parameters [o = oA] are close to the average global parameter of 1.50 for 47Ag and the 50Sn-42Mo pair. However, the values of o are considerably larger for 48Cd and 43Tc-49In pair, implying that these radionuclides could be formed by more than one process. The data for 48Cd and the 43Tc-49In pair can be represented better by two Gaussian functions (symmetric and asymmetric), each with a o value of 1.50 and peak separation of about 4 A’ (Fig. 14). Significant deviations occur for the curve of the 43Tc-49In pair, underlining the need for a re-evaluation of these data.

Worldwide, LWRs (PWRs, BWRs and WWERs) are the major types of nuclear power plants. They represent approximately 88% of today’s global nuclear power capacity, and evolutionary designs, based on this experience base, are being developed in several countries. The major evolutionary LWR designs are shown in Table V.

The evolutionary LWR activities in different countries are briefly described in the following10:

In the USA, designs for a large sized advanced PWR (the Combustion Engineering System 80+) and a large sized BWR (General Electric’s ABWR) were certified by the U. S. NRC in May 1997. Westinghouse’s mid-size AP-600 design with passive safety systems was certified in December 1999. Efforts are currently underway by Westinghouse on a 1090 MWe plant called the “AP-1000,” applying the passive safety technology developed for the AP-600 with the goal to reduce the capital costs through economies-of-scale. A certification application for the AP-1000 design has been made to the US NRC this year. General Electric is also designing a 1380 MWe ESBWR applying economies-of-scale together with modular passive safety systems. The design draws on technology features from General Electric’s ABWR and from their earlier 670 MWe simplified BWR with passive systems.

In France and Germany, Framatome ANP completed the basic design for a 1545 MW(e) European Pressurized Water Reactor (EPR) in 1998, which meets European utility requirements. The EPR design includes the mitigation of core melt and vessel penetration accident scenarios ensuring the avoidance of evacuation of people in the vicinity of the plant. Accidents with molten core material outside the reactor pressure vessel are handled via a spreading concept in the basement of the containment. The EPR’s higher power level relative to the latest series of PWRs operating in France (the N4 series) and Germany (the Konvoi series) has been selected to capture economies of scale. Framatome ANP’s SWR 1000 is based on German BWR experience with added features to increase safety. It is an advanced BWR with active and passive safety features which allows for extended grace period for accident control and consequences of a core melt accident is limited to the immediate vicinity of the plant. This has been achieved by providing cooling of the reactor pressure vessel exterior. The essential elements of the SWR safety concepts are shown in figure 4.

SWR 1000 Safety Concept

In Sweden, Westinghouse Atom is also developing the 1500 MWe BWR 90+, an advanced boiling water reactor with improved safety and operability. This is an upgraded version of the BWR operating in Sweden and Finland.

The first two ABWRs in Japan, the 1360 MWe Kashiwazaki-Kariwa 6 and 7 units, have been in commercial operation since 1996 and 1997, respectively. ABWR plants are under construction at Hamaoka Unit no. 5 and Shika Unit no. 2, and under licensing at Ohma Unit no. 1. Another eight ABWR plants are in the planning stage in Japan. The benefits of standardization and construction in series are being realized with the ABWR units. Expectations are that future ABWRs will achieve a significant reduction in generation cost due to standardization, design improvements and better project management. In addition, a development programme was started in 1991 for 1700 MWe ABWR-II, aiming to further improve and evolve the ABWR, with the goal of significant reduction in power generation cost. Commissioning of the first ABWR-II is foreseen in the late 2010s. Also in Japan, the basic design of a 1530 MWe advanced PWR has been completed by Mitsubishi Heavy Industries and Westinghouse for the Japan Atomic Power Company’s Tsuruga-3 and -4 units.

In the Republic of Korea, the benefits of standardization and construction in series are also being realized with the 1000 MWe Korean Standard Nuclear Plant (KSNP). The first two KSNPs, Ulchin 3 and 4, have been in commercial operation
since 1998 and 1999, respectively, and four more units (Yonggwang 5 and 6 and Ulchin 5 and 6) were under construction in 2001, with Yonggwang 5 and 6 scheduled to begin commercial operation in 2002. In addition, ROK is developing the Korean Next Generation Reactor, now named the Advanced Power Reactor 1400 (APR- 1400), which is focusing on improving availability and reducing costs. It has received design certification and is expected to be constructed by 2010.

In the Russian Federation, efforts continue on evolutionary versions of the currently operating WWER-1000 (V-320) plants. This includes the WWER-1000 (V — 392) design, of which two units are planned at the Novovoronezh site, and WWER — 1000 units are also planned in China, India and the Islamic Republic of Iran. Development of a WWER-1500 design has been initiated. Development is also ongoing on a mid-size WWER-640 with passive safety systems, and on an integral design with the steam generator system inside the reactor pressure vessel.

In China, the China National Nuclear Corporation (CNNC) is developing the CNP-1000 plant. China is pursuing self-reliance both in designing the plant to meet Chinese safety requirements, and in fostering local equipment manufacture with the objective of reducing construction and operation costs. Lessons learned from the design, construction and operation of the Qinshan and Daya Bay NPPs are being incorporated. Two ABWRs are under construction in Taiwan.